13 Study .Abroad Research: Findings, Implications, and Future Directions

JOSEPH COLLENTINE

One of the most important variables that affects the nature and the extent to which learners acquire a second language (L2) is the context of learning, that is, whether the learning takes place within the society in which the L2 is productive or where the first language (Ll) is productive. Second language acquisition (SLA) research takes place in three primary contexts, which differ in sociological and nmctional terms. The foreign-language (FL) classroom exists in the domestic setting of the LI, and learners tend to use the L2 within the classroom and as it relates to academic purposes. The intensive domestic immersion setting is different from the FL context in that students dedicate the majority (if not all) of their academic term to Shldying the L2, and this context often entails an increase in the nmctional purposes of the L2 when learners sign a "contract" not to use the LI. The study abroad (SA) context takes place lll. c01.mh"ies where the L2 enjoys an linportant sociological and nmctional status, entailing a combination of plaIUl.ed curriculum and a host family. This chapter provides an overview of the state of the art of SA research. This review 1.mcovers a pattern not clearly articulated to date: specifically, the most salient domains of interest (e.g., cognitive, pragmatic, sociolinguistic factors) that arise out of the SA literature are quite different from the salient domains in other contexts of learning. Informed by this literature, I conclude with recommendations as to the ideal design of SA programs in a modern FL curriculum.

It is extremely challenging for researchers to isolate the effects of the learning context on acquisition because one must be aware that, within any of these sihlations, learners acquire the L2 lll. two sub-contexts: communicative contexts as. well as learning contexts (Bats tone, 2002). Communicative contexts require that the learner use the L2 to exchange information and engage in essential social and interpersonal functions. Learning contexts manage input and output so that learners will attend to form and take intentional steps toward linproving their linguistic expertise. The influence of these covariates complicates the assessment of a leaJ:ning context since the FL classroom heavily favors learning contexts, intensive domestic immersion settings attempt to provide both comm1.micative and learning contexts, and SA presumably provides more opporhmities for

rch: lS,

is

Ie and the extent to d of learning, that is, 1 the L2 is productive age acquisition (SLA) :1' in sociological and n. the domestic setting )m and as it relates to , is different from the ~1l) of their academic s an increase in the lct" not to use the L1. lere the L2 enjoys an mbination of phw,ed 'erview of the state of : clearly articulated to , cognitive, pragmatic, :e quite different from led by this literature, of SA programs in a

effects of the learning , within any of these nmunicative contexts as ~ contexts require that age in essential social put and output so that lward improving their ::>licates the assessment ~ors learning contexts, e both communicative :lore opportunities for

Study Abroad Research 219

processing and using the L2 in commmucative contexts. SA learners must deter­mine the relationship between the L1, the L2, and their identities as social individuals and language learners. For researchers, the complication then becomes which theoretical frameworks can capture the cognitive and social developments through which learners pass in the SA setting. As Collentine and Freed (2004) note, research in a SA context provides an important contextualization for lmder­standing the interaction between cognitive, sociolinguistic, and socio-cultural factors in the construction of a comprehensive theory of SLA. This chapter attempts to delineate the contribution of SA research to SLA theory by exploring its (brief) history as a sub-discipline. The chapter also details the types of populations that researchers have studied, the efficacy of tlus learning context, how researchers operationalize "efficacy" (and the effects on the interpretation of results), the roles of input and interaction, the cOgIutive changes (e.g., phonolog­ical memory) that occur abroad, important issues identified in the research relating to learner identity, and the role of pedagogy in SA contexts on acquisition.

A Brief History of Study Abroad Research

SA research can be seen as having two periods. The first attempted to lmder­stand the overall efficacy of SA programs. These studies concentrated on measuring the gains learners make abroad largely from broad measurement instruments. This period extends from the 1960s to Barbara Freed's publication of her seminal volume Second language acquisition in a study abroad context in 1995 (Freed, 1995b). Freed succeeds in framing SA research within the SLA theory­building enterprise, challenging researchers to view SA research as a means of studying the effects of "learning context" on acquisition.

The first period examined gains (or simply post-treatment abilities) with instruments that sought to assess learners' overall L2 abilities.1 Carroll's (1967) widely-cited study looked at the language skills of 2,782 college seniors on tests that measured linguistic skills in the L2 (i.e., their metal:inguistic knowledge), finding that even a short duration abroad (touring or summer) predicted higher levels of proficiency. Willis et al. (1977) summarized a series of studies on British students, concluding that these studies lacked an overall systematic assessment of learners gains. Willis et al.'s (1977) own study on British shldents in France and Germany showed general support for residency abroad. The strongest gains were in listening and speaking and less in reading abilities. Dyson (1988) con­ducted another macro study OD 229 British shldents in. France, Germany, and Spain, showing that the learning context improved listening and speaking skills. Opper, Teichler, and Carlson (1990) conducted a large-scale Shldy on the efficacy of SA on students in more than 80 programs in Britain, France, Germany, and Sweden. The Shldy is limited in its validity because it relies on self-reported assessments of general language skills (listening, reading, speaking, and writing) whereby the learners generally reported important gains. It proVided nonetheless an important clue to a key factor that has been identified in the SA research: SA appeared to be particularly powerful for learners with lower levels of proficiency.

220 Joseph Collen tine

Perhaps the key observation to be gleaned from this study is the notion that there are threshold levels of development at which SA will be optimally beneficial (discussed further below). Another key ShIdy of this period that laid the fmmda­tion for current SA research was Mahle and Raupach (1983), who fmmd SA not to have an important effect on improving morphosyntactic abilities but to have a positive effect on fluency, a factor in SLA research that has become an important focus of late since it helps us to lmderstand the interaction between online pro­cessing mechanisms and linguistic competence such as working and phonolo­gical memory (see Segalowitz, 2003). The final noteworthy shIdies of this period are Brecht and Davidson (1991) and Brecht, Davidson, and Ginsburg (1995), who examined 668 American learners' acquisition of Russian in a SA context. The data reinforced a growing hunch that individual differences (e.g., reading aptihIde) are exceptionally evident in this learning context and that preprogram explicit grammar instruction predicts gains abroad.

Freed (1995a) is the first effort to synthesize SA research. She identifies a number of issues that researchers in her volume and others have addressed since its publication. She notes a growing suspicion that the linguistic benefits of the SA context were not the same as those of the traditional classroom (Regan, 1995), and there were surprisingly few empirical studies that actually compared these two learning contexts. She also recognizes the hypothesis that there might be a proficiency threshold at which learners most benefit from SA (Brecht & Davidson, 1991; Brecht, Davidson, & Ginsburg, 1995; Regan, 1995). Additionally, there was almost no research that examined SA gains within current theoretical frame­works that dealt with how new linguistic information becomes internalized in the learner's competence (e.g., interactionist, input-oriented, and socio-cultural models). Freed (1995b) presents shIdies addressing two issues in her volume, providing important SA-AH (at horne) studies comparing orders of acquisition abroad to orders documented in FL contexts (Guntermann, 1995; Lafford, 1995). Freed (1995a) also provides shIdies on the acquisition of pragmatic competence within a sociolinguistic framework (Regan, 1995; Siegal, 1995).

Since Freed (1995b), noteworthy collections of SA experiments have been published. And, while these collections have concentrated on Americans going abroad (Collentine & Freed, 2004; DuFon & Churchill, 2006; Gore, 2005; Pellegrino­Aveni, 2005), Murphy-Lejeune (2002) examines SA in the Emopean context. What follows delineates the key topics that SA research is currently addressing. The research addresses a variety of issues that relate to the internal cognitive mech­anisms affecting acquisition and the external sociolinguistic mechanisms, as weLl as the socio-cultural issues of SLA in a SA context of learning.

Populations of Study and the Threshold Hypothesis

Kinginger (2007), as well as Coleman (1997), characterizes the existing body of SA research as falling into two categories, with each focusing on different populations and distinct levels of development. Research on American, university-level learners tends to examine acquisition at the begimung stages of

( is'the notion that there be optimally beneficial od that laid the founda-183), who found SA not ic abilities but to have a 'lS become an important on between online pro­working and phonolo­

ly shldies of this period , and Ginsburg (1995), Llssian in a SA context: ifferences (e.g., reading xt and that preprogram

She identifies a l1lunber lve addressed since its .listic benefits of the SA assroom (Regan, 1995), ctually compared these s that there might be a 3A (Brecht & Davidson, Additionally, there was Tent theoretical frame­Jecomes internalized in lted, and socio-culhual , issues in her volume, 19 orders of acquisition fi, 1995; Lafford, 1995). pragmatic competence

1995). :xperiments have been :d on Americans going ; Gore, 2005; Pellegrino­Iuropean context. What :rently addressing. The lternal cognitive mech­tic mechanisms, as well :ning.

1 Hypothesis

es the existing body of focusing on different

:search on American, the beguming stages of

Study Abroad Research 221

development, with shldy participants sampled from first- and second-year language programs. While there is a good amount of research on learners at the beginning levels of development, Lafford and Collentule (2006) report that SA research on American students of Spanish has sampled participants whose preprogram proficiency ranged from the novice to the advanced high levels. It is true that there exists little information about American learners' success Ul advanced-level, direct-enrollment programs (i.e., where students sit in classes with otherwise proficient/native speakers of the L2). American, university-level L2 programs - and so SA curricula - concentrate on fostering acquisition at the financially lucrative early stages of acquisition (where FL enrollments in general are highest). Kinginger (2007) asserts that SA research such as Murphy-Lejeune (2002) has focused on European learners at more advanced levels, stemming from the European Commission's ulter-1Uliversity programs, such as the ERASIVfUS program, which allows students to complete part of their lUliversity shlclies in another ED cOlUltry and lU1iversity (41 percent of such students Shldy language or philology; http://www.erasmus.ac.uk/statistics/contents_04_05.htm) .

There is a growing interest in SA SLA issues along the Pacific Rim, and these learners - Japanese, Chinese, Korean, and Taiwanese university-level adults slightly beyond the initial stages of development - tend to target English as their L2, either in North America or the South Pacific (Churchill, 2006; Tanaka & Ellis, 2003). If current estilnates that 80 percent of foreign shldents in the world are from Asian cOlUltries are correct (Altbach & Bassett, 2004), these populations will become an ulcreasingly important source of data.

There are ilnportant pockets of SA contexts and learner profiles that have yet to be studied. Shldents are attendulg so-called language camps with increasing frequency. Korean.net, for ulstance, reports that nearly 40,000 Korean shldents enrolled in domestic EFL immersion programs in 2005? The New York Times recently reported that nearly 75 percent of American summer camps have foreign nationals attending their activities, and many camps provide some EFL instruc­tion (Bick, 2007). To this researcher's knowledge, there is no existing literahue either documenting these experiences or providing data on their value, although the usuallUlsubstantiated and anecdotal claims of efficacy are not hard to lU1cover.

Learners of all levels of development (who are largely self-selecting) have been shldied in the SA literahue. However, Lafford and Collentine (2006) discuss the growing consensus amongst researchers that there is a threshold which learners must reach to benefit fully from the SA context of learning. And, while the general notion of a threshold level is unportant for program designers to keep in mind, from a linguistic compet~nce perspective it is probably too broad in scope. There are most likely specific domains that require a particular developmental threshold for overall gains to occm. Golonka (2006) - refinillg the Brecht, Davidson, and Ginsberg (1995) analysis - presents evidence suggesting that preprogram linguistic (grammar, vocabulary, accuracy) and metalinguistic (self-corrected errors and sentence repair) levels predicted which SA learners of Russian would attain the Advanced level on the ACTFL proficiency scale. Segalowitz and Freed (2004) fOlUld that, amongst Spanish L2 learners, an initial threshold level of basic word recognition and lexical access processulg abilities may be necessary for oral

, 1

I

222 Joseph Collen tine

proficiency and fluency to develop abroad significantly. Lafford (2004) surmises that advanced learners bring more formulaic expressions to SA communicative contexts and can therefore spend more attention reSOUTces on form, whereas novices must attend primarily to meaning. Finally, Segalowitz and FTeed (2004) report that the most important gains that SA learners of Spanish make abroad are in the domain of fluency (as meascued by temporal/hesitation phenomena), and O'Brien et al. (2007) report that these fluency gains abroad are a ftmction of students' potential for phonological memory storage, which varies from adult to adult. While this last factor is not controllable from a preprogram perspective, it argues that a shldent's cognitive and linguistic abilities will mediate develop­mental gains in the program.

The Issue of Study Abroad Efficacy

Many researchers and educators have surmised that the SA context is the sine qua non for achieving global L2 competence (Rivers, 1998). Freed (1995a) acknow­ledged that, while some empirical studies conducted up cmtil 1995 provided evidence that SA facilitates acquisition (e.g., DeKeyser, 1991; Teichler & Steube, 1991), a sizable amolmt of evidence challenged researchers to consider whether the Shldy abroad context might impede acquisition at the beginning stages (e.g., Freed, 1990; Spada, 1985, 1986), which, again, points to a threshold effect. Cohen and Shively (2007) summarize the research to date, asserting: "An intriguing finding in the study abroad research litera hue that prompted the current Shldy was that study-abroad shldents do not necessaTily achieve greater language gains than their peers who stay home and study the target language" (p. 189). Clearly, overall efficacy is difficult to assess in the absence of data stemming from large­scale shldies such as those reported by Brecht and Davidson (1991) and Segalowitz et al. (2005). Yet, SA studies tend to be longihldinal in nature, as opposed to much SLA research that examines short-term effects. These studies support Carroll's (1967) initial assertion about the advantages of SA, but they require a great deal of qualification: whereas SA affects gains in certain language-specific domains, it does not affect development in all aspects of a learner's competence. Interest­ingly, linguistic aspects that do indeed seem to benefit from SA, such as fluency and discursive abilities, are often not those in which AH FL program directors hope to see improvements, such as those grammatical aspects around which the AH, focus-on-forms syllabus is designed. The following two sections summarize what we know to date about the important cognitive and linguistic aspects of L2 development with which SA interacts.

Important Cognitive Constructs in the Study Abroad Literature

One consistent theme in the SA literahlre is individual differences and lll.dividual variation (variation is addressed in the next section). An area of SLA research

Hord (2004) surmises o SA comnnmicative :s on form, whereas ritz and Freed (2004) :ush make abroad are lon phenomena), and ad are a nmction of l varies from adult to ogram perspective, it 'ill mediate develop-

context is the sine qua eed (1995a) acknow­until 1995 provided 1; Teichler & Steube, to consider whether

eginning stages (e.g., reshold effect. Cohen 'ling: "An intriguing :ed the current study reater language gains 1ge" (p. 189). Clearly, temming from large­)991) and Segalowitz ,as opposed to much ies support Carroll's T require a great deal e-specific domains, it :ompetence. Interest­"1 SA, such as fluency ~L program directors cts ar01md which the ) sections summarize :lguistic aspects of L2

udy

~ences and individual lTea of SLA research

Study Abroad Research 223

that is making important contributions to our tmderstanding of individual dif­ferences - or, between-subject variation - examines the interaction of short-term memory stores (e.g., working memory, phonological memory), speed of informa­tion access, attention control, and acquisition (see DeKeyser, 2001).

Segalowitz and Freed (2004) are interested in the cognitive prOCeSSlll.g abilities that LUl.derlie "expert abilities." They present data indicating that cognitive abilities interact with development abroad in complex ways. They fmmd lexical access to be related to overall proficiency gains amongst SA learners of Spanish, such that learners who access lexical items faster show greater gains. They also present data suggesting that learners exhibiting greater attention control at the com­pletion of their SA program spoke less fluently. Segalowitz and Freed (2004) speculate that the increased attention control in the SA context may reflect increased monitoring of output. O'Brien oversees two studies examining the interaction of phonological memory and SA gains in learners of Spanish, both of which indicate that phonological memory assists not only children (in voc­abulary gains) but also adults in an L2 context. O'Brien et al. (2006) showed that phonological memory abilities have a positive effect on one's abilities to produce multi-propositional utterances at the early stages of Spanish learners' development (i.e., narrative abilities) in addition to a positive effect on the acqui­sition of particular grammatical hmctors at later stages. O'Brien et al. (2007) add to our tmderstanding of the cognitive mechanisms underlying the general fluency gains associated with SA experiences, showing that phonological memory also predicts fluency gains. Finally, Tokowicz, Michael, and Kroll (2004) examine working memory capacity and single-word translation errors, presenting data that suggest that SA learners engage in more approximate than precise transla­tions but that this tendency is lilnited to those with higher working memory capacities.

This literature suggests that learners' predisposed cognitive abilities determine how much one can produce (and how fast) as a result of SA, and may impact the ways that learners approach attention-demanding processes (such as transla­tions and story-telling). It also indicates that researchers ought to consider carehllly how they interpret observable behaviors, such as fluency. The Segalowitz and Freed (2004) study provides some evidence that where there is a lack of fluency, there may be more monitoring occurring (due to greater attention control), which is not an tmreasonable conclusion, since Golonka (2006) shows a strong relationship between metalinguistic abilities and overall proficiency gains.

Important Linguistic Constructs in the Study Abroad Literature

As mentioned above, there has been a growing concern about the overall lack of efficacy of SA on acquisition (d. Cohen & Shively, 2007). In the following I show that, while it is not accurate to claim a superiority for SA on fostering acquisition, it is likewise erroneous to conclude that important, positive changes

224 Joseph Collentine

(toward native-speaker norms) do not occur as a result of the SA context of learning. The developmental growth one finds in this learning context appears not to be what one finds in the AH setting (although the field admittedly has a dearth of comparative AH-SA shldies) nor what learners and theu· AH teachers expect (e.g., dramatically improved grammatical and phonological accuracy). The linguistic-specific growth that occurs is in the domain of global discursive abilities, the expansion of knowledge of verb paradigms, and in the domain of pragmatics. A review of this research reveals much about the commlmicative needs and demands placed on learners in the SA context.

One of the consistent conclusions that researchers have drawn from the SA literature has been that the development of grammatical abilities does not seem to outpace that of the AH context. DeKeyser (1990, 1991) fOlmd that residence abroad had little impact on the development of overall grammatical abilities and that SA learners were equal to or inferior to their AH counterparts in their use of grammar. Collentine (2004) gauged SA learners' acquisition of a variety of morphosyntactic features, showing that they do not make as much progress as AH learners on precisely those grammatical aspects that many FL teachers emphasize, namely, verbs and subordill.ate conjunctions.

These shldies, ill. sum, ill.dicate that the appreciable development of morpho­syntax and general grammatical abilities is not to be expected, at least within the timeframe of a semester to a year abroad. Indeed, two of these studies (Collentine, 2004; DeKeyser, 1990) suggest that the AH experience affords certain advantages as regards overall grammatical development for intermediate learners. A notable exception is Isabelli ml.d Nishida (2005), who reveal that SA has an advantage with respect to subjlmctive development when learners are at more advanced stages, thus supporting the threshold hypothesis.

A further complication to this scenario arises out a comparison of Golonka (2006), who fOlmd preprogram metalinguistic knowledge to predict proficiency gains, and Izumi and Iwasaki (2004). The latter examined the effects of amount of SA experience on a grammaticality judgment test by Japanese-speaking English FL learners, where the participml.ts were asked to give reasons for their evaluations. The purely classroom EFL learners used intuitive, analyzed, and metalill.guistic knowledge more or less equally, while the learners who expe­riencedliving abroad for several years used ul.tuitive knowledge and very little metalinguistic knowledge. It may be that preprogram metalinguistic knowledge is an important prerequisite, but that not all learners depend on that knowledge in communicative situations to process grammar. It may also be that, as the Segalowitz and Freed (2004) data suggest, some processes become more auto­matized abroad, so that metalinguistic awareness is converted mto implicit grammatical knowledge. Clearly, more research is needed m this area.

Research is starting to suggest that the orgmuzmg principles arOlmd which SA learners develop their grammatical abilities stem from the discursive demands they face m commlmicative contexts. This ought not to be surprising, given what we have known for a while about uninstructed SLA The Second Language Acquisition by Adult Immigrants shldy was conducted from 1981 to 1988 in five

of the SA context of rning context appears field admittedly has a and their AH teachers l.onological accuracy). n of global discursive and in the domain of

ut the communicative

e drawn from the SA lbilities does not seem ) found that residence

grammatical abilities { counterparts in their cquisition of a variety cake as much progress hat many FL teachers

velopment of morpho­:ted, at least within the ese studies (Collentine, rds certain advantages .ate learners. A notable SA has an advantage

are at more advanced

)mparison of Golonka to predict proficiency

l the effects of amOlmt by Japanese-speaking d to give reasons for ed intuitive, analyzed, the learners who expe­>wledge and very little talinguistic knowledge ~nd on that knowledge ly also be that, as the es become more auto­onverted into implicit 1 ill. this area. iples arOlmd which SA le discursive demands surprising, given what The Second Language

m 1981 to 1988 in five

Study Abroad Research 225

European countries by the Max Planck Institut far Psycholinguistik, mostly on L2 learners of German (Perdue & Klein, 1992), showing that L2 grammaticaliza­tion arises out of functional pressures to achieve discursive coherence. Collentine (2004) used corpus-based techniques (see Biber, 1988) to compare the develop­ment of narrative abilities, as well as growth in the semantic density, of AH and SA learners of Spanish, revealing that the SA context afforded a significant advantage. Similarly, Cheng and Mojica-Diaz (2006) compared SA learners of Spanish on their subjunctive abilities (using a native-speaker baseline), finding no significant improvement after a two-month period, although they report some learners started producing more tightly structured argumentation over time.

Fine-grained analyses of SA learners' grammatical performance over time sug­gest another plausible explanation for why researchers in the past have not fOlmd an appreciable quantifiable advantage for SA context in granunatical develop­ment: the SA context fosters grammatical variation. Regan (1995) examined SA learners of French and their use of the ne negation particle, showing that AH learners tended to adopt a single, standardized construct for negation whereas their SA c01.mterparts varied between the inclusion and omission of the ne particle as a function of sociolinguistic factors. Howard (2002, 2006), who has examined SA Irish learners of French, has presented convincing evidence that SA learners spend much time varying within and accommodating the individual elements of a paradigm (e.g., present-tense French inflections). His data also suggest that L2 phonological development interacts with the increased inflec­tional variability (or perhaps inflectional confusion) in the SA context, which is not surprising given that inflectional morphology tends to comprise short phon­emic segments (and suffers neutralization in modern French; d., Howard, Lemee, & Regan, 2006).

Some researchers suspect that one of the reasons for a weak SA grammar effect is that learners have much less access to the L2 than one might suspect, especially as it relates to its pragmatic feahl.Yes. Native speakers find it difficult to abandon certain mentor-apprentice modes of interaction with learners (Pellegrino-Aveni, 2005; Wilkinson, 2002). Barron (2003), examining Irish SA learners of German, argues that learners form social networks with other speakers of theiJ: Ll, limit­ing their access to native speakers. Others suggest that learners are not adequately aware of the rules of pragmatics within. the target culhl.Ye (Wilkinson, 2002). Cohen and Shively (2007) present data from a controlled experiment suggesting that preprogram efforts to raise French and Spanish learners' awareness of speech acts had only a marginal effect on their ability to mitigate the intensity of requests and no significant effect on their ability to recognize the appropriateness of other acts, such as apologies.

Some recent research on SA pragmatics reinforces O'Brien et a1.'s (2006) posi­tion that SA has differential effects on learners depending on their preprogram level of development.3 Shardakova (2005) reports that SA learners of Russian with low preprogram proficiency adopt culhlrally appropriate apologies, while more advanced candidates develop their own strategies, which are not consistent

226 Joseph Collen tine

with cultural prescriptions, reflecting the extent to which linguistic behaviors in a SA environment can be a hmction of identity as much as input and types of interaction (see below for more on the notion of identity). This pattern of tension between pragmatic appropriateness and rejection of target cultural (linguistic) practices in terms of routines and speech acts is a recurring theme in the SA literature (DuFon & Churchill, 2006). This may account for the general consensus amongst researchers of SA pragmatics that, while SA learners outpace their AH cOlmterparts in pragmatic development, they may spend years attaining native­speaker behaviors, if they so choose. All told, pragmatic competence seems to develop quite slowly in the SA context (Hoffman-Hicks, 1999; Rodriguez, 2001). Protracted pragmatic development is not surprising since learning new scripts/ discourse grammar (i.e., in the case of speech acts) and illocutive meanings along­side locutive ones (i.e., double meanings for certain phrases and constructs, such interrogatives that represent imperatives) represents a considerable task.

This review indicates that grammatical and pragmatic development abroad becomes complicated by the sociocognitive and socio-cultural pressures that learn­ers face in the SA context, a sihmtion that sends many more messages to learners than does the AH context as to the complete repertoire of skills and behaviors one needs to be communicatively functional. Communicative demands at the discursive, as opposed to the sentential, level may well force the learner to make adjustments to his or her preprogram, personal or internal syllabus (Lafford, 2004). If linguistic variation is more prevalent in the SA context than the AH context, SA learners may need to incorporate non-standard forms into their development competence. This greater variability in the SA input may force the SA learner to accommodate the L2 forms with which AH learners may only need to (re)familiarize themselves at test time. Finally, pragmatic development seems to be slow in the SA context, yet important gains are made in the SA context. However, issues of (self-) identify may well interact with the extent to which pragmatic norms become adopted by the learner.

The (Assumed) Roles of Input and Interaction

One of the most undisputed assumptions about the SA context is that learners receive vast amOlmts of input and have numerous opportunities for commun­icative interaction. There is a tendency in the field to ath'ibute gains that learners make to the enormous allOlmt of available input. However, there has been no attempt independently to document in a fully quantified manner the types of input and interaction that learners have abroad, The assumptions that exist in the literature may be too strong. Pellegrino-Aveni (2005) documents from a qualit­ative perspective the types of interactions SA learners of Russian have, arguing that self-preservation (e.g., face-saving) needs effectively impede learners' contact with native speakers, There have been attempts to document the types of input and interaction learners have in a SA context, such as the Language Contact Profile (LCP) employed by Segalowitz and Freed (2004). This sort of

19uistic behaviors in a lS input and types of 11is pattern of tension ~t cultural (linguistic) :ULg theme in the SA the general consensus lers outpace their AH 'ears attaining native­competence seems to )99; Rodriguez, 2001). learning new scripts / _ltive meanings along-3 and constructs, such 3iderable task. development abroad

11 pressures that learn­~ nLessages to learners f skills and behaviors ltive demands at the :e the learner to make Lal syllabus (Lafford, context than the AH lard forms into their \. ULput may force the arners may only need c development seems de UL the SA context. I the extent to which

lction

mtext is that learners tunities for COlTIrrllID­lte gains that learners er, there has been no manner the types of

)tions that exist in the unents hom a qualit­~ussian have, arguing ~ly unpede learners' ) document the types uch as the Language i (2004). This sort of

Study Abroad Research. 227

assessment tool is helpfut in that it has built-in redlmdancies, so as to triangulate the reported contact for any given subject. However, it is a self-report, and its validity is limited in the same way all self-reports are (such as in clinical research), where data sets represent self-perceptions and require objective, third­party validation, such as by host families and professors abroad.

TILe lack of prilnary data on the amolIDt of input learners receive abroad is, in fact, symptomatic of a larger problem in SA research and in the enterprise as a whole. Large-scale shldies of how much input and interaction learners have in AH contexts really do not exist either. Yet, the field of SLA knows a good deal about the effects of input and interaction on acquisition because we have almost three decades of controlled treatment data (Mackey & Goo, 2007; Norris & Ortega, 2001). No concerted research agenda exists either in theory or practice to control the amOlmts and types of input and interaction learners receive in the SA context. Taking this observation a step further, we are confronted with the conclusion that we do not know how input differs in the SA context, comprehens­ible or otherwise. Barron (2003), for instance, notes that learners often misinter­pret ilnportant aural cues and draw erroneous conclusions about the L2, since negative evidence is not available. Additionally, Magnan and Back (2007), using a modified version of the LCP, present data suggesting that the living situation and access to authentic aural media do not predict oral proficiency gains abroad. The field of L1 acquisition has developed a number of tools for documenfuLg what learners are exposed to in nahll'alistic settings, and SA researchers could use these methods and instruments as a startulg point.

McGeeking (2006), shldying SA students of Japanese, is one of the few shldies to document how learners negotiate for mearting in the home-stay environment, reporting that numerous opportunities exist (see also Dings & Jobe, 2003, as well as Smartt & Scudder, 2004). Yet the opportunities to negotiate appear to be mitig­ated in the SA context by personal and interpersonal factors. Wilkinson (2002) reports that home-stay families find it difficult to use nahll'alistic language with their SA learners of French, preferruLg to use teacher-talk, denying the learners opporhlnities for authentic input via interaction. Churchill (2006) documents that SA learners of English from Japan vary in the amOlIDt of authentic uLteractions they have depending on length of stay, which in turn is a function of the extent to which the learners are fully integrated into the target learning commmtity, which is difficult in five-week programs.

What is lmclear is whether there are more reports that have made it to press in the SA literahlre than in the AH literahll'e about the interpersonal mitigating factors that impede opportunities for negotiatuLg for meanuLg. Much research purporting to comment on negotiation opporhrrtities is conducted within a socio­culhlral hamework (d. Kinginger, 2007; DuFon & Churchill, 2006), and so the epistemology is not the same as that from which SLA has traditionally examined interaction (Le., hom a largely cognitive perspective; McGeeking (2006) is a notable exception). Until these two epistemologies can achieve a common terminological interface, our lIDderstanding of negotiation in SA contexts relative to the existing SLA literahlre will remain weak.

228 Joseph Collentine

Identity

Researchers approaching the SA context from a socio-cultural perspective have focused on the individual histories of students and the tension that exists between maintaining individuality, issues of self esteem, worldviews (e.g., social hierarchies), and the need to advance their own development through native­speaker interactions (Wilkinson, 1998). Churchill (2006) documents how the man­ner in which learners are received (or not) affects the amOlmt of interaction they have. Kinginger (in press) provides an exhaustive sunmlary of the socio-cultural literahlre relating to SA. She concludes that the most important theme is that, in those moments when learners' personal sense of identity with the target culture (or with the representatives that they know, such as the home-stay family) is distant, or when learners sense that they have too many obstacles to attaining higher levels of proficiency (either through negative interactions or from a sense of linguistic inadequacy), SA learners abandon their role as "language learners," thus impeding the development process.

Programmatic Considerations

There are a number of programmatic implications to be gleaned from the research presented to date. From an administrative perspective, it is difficult for a home institution to have a strong effect on a host institution's syllabi and methodological approaches. SA instructors are often employees of other instihltions and adjlmct faculty who have less of a stake in the long-term needs of the learner or program. Programs such as the University of Delaware's, where there is a tight integration between the home and abroad curricula and a faculty development program, can serve as models for ensuring greater quality control over pedagogy (d. Chieffo & Zipser, 2001). The last ten years have seen a notice­able interest in, and research on, the role of tasks and authentic interactions within an L2 curriculum. For instance, Doughty and Long (2003) suggest that Task-Based Language Teaching provides one such curricular framework that, based on SLA research, details the types of interactions in which learners can engage according to developmental level. An exploration of the plausibility and the outcomes of this sort of language program within a SA curriculum might at least provide a principled starting POlll.t from which to Shldy the "missing SA methodology" (Lafford & Collentine, 2006).

It is lmclear what the ideal duration of a SA program might be, although a consideration of this question raises intriguing questions for SLA researchers. Programs tend to range from about five weeks to a year in length. Expectations about (1) how much development occurs and, more importantly, (2) which aspects of a learner's competence develop must be measured against the observa­tions made above that exponential growth will not occur within such a time frame. In the absence of a solid base of SA-AH comparative shlClies, it is possible

ultural perspective ~ tension that exists ldviews (e.g., social :nt thTough native­nents how the man­t of interaction they )f the socio-cultural mt theme is that, in h the target cui hue Jme-stay family) is lstacles to attaining ons or from a sense language learners,"

gleaned from the ctive, it is difficult institution's syllabi ,mployees of other he long-term needs : Delaware's, where ricula and a faculty ater quality control have seen a notice­thentic interactions (2003) suggest that ar framework that, which learners can the plausibility and urriculum might at ly the "missing SA

ight be, although a )r SLA researchers. 2ngth. Expectations Jrtantly, (2) which 1.gainst the observa­Nithin such a time tudies, it is possible

Study Abroad Research 229

(naively) to look at SA research solely as a set of longitudinal SLA investigations. This would lead SLA researchers to the conclusion that L2 acquisition is a pro­tracted process. If the field were to imagine (again, naively) that AH research represents the possible effects of pedagogical interventions in a laboratory setting, SA research indicates that sociopragmatic variables and the inherent linguistic variation existing outside this laboratory will essentially increase the "content" that learners must acquire. The SA learner is confronted with hmc­tional demands that invite us to view the AH student as one who learns the language for so-called special purposes. As Freed (1995b) notes, it is no longer tenable to consider a student who enrolls in a SA program as a FL learner, since the SA converts him/her into the learner of a L2. Thus, the end goal of the SA shldent (whether or not the learner completely abandons his/her role as a lan­guage learner; d. Kinginger, in press) is often different from the AH learner's, and the amolmt of time needed to complete the L2 agenda becomes longer than that needed for the FL agenda.

If there is a consensus, it is that a student may begiIl a SA program too early in his/her development. The research considered above indicates that a certain level of metalinguistic knowledge (we do not know yet how much) is a prerequi­site for developing the L2 abroad (Brecht & Davidson, 1991; Magnan & Back, 2007). At the very least, SA outcomes are sensitive to learners' preprogram com­petence levels (Golonka, 2006; O'Brien et al., 2006). At this point, it is possible to suggest: (1) the amOlmt and type of preprogram preparation should be studied in greater depth; (2) these preprogram considerations should inform the types of L2 knowledge and levels of development in programs.

The research does not control for the effects of the home stay with host families. A review of the litera hue indicates that most students stay with a host family. Lazar (2004), however, reports that the achml amOlmt of time that learners spend with their host families varies both in quantity and quality, and these interactions have an appreciable effect on acquisition in general. Lafford (2004) found a significant negative correlation between the amolmt of time spent talking with host families and the use of conununication strategies to bridge communication gaps. The literature reported here indicates that issues of (1) identity and (2) the host family's perceived role as a mentor may determine how much one learns (Wilkinson, 2002).

Concluding Remarks

Most research conducted to date on SA has concentrated on the American university experience, as seen by the above literature review. As Kinginger (2007) notes, other parts of the world such as Europe have a different relationship with the SA experience, depending on their proximity to other languages and the value they place on bilingualism. Perhaps because the field of SLA is still in the early stages of building a theory of acquisition, it may not be surprising that very little research exists on the effects of SA on learners such as those in the ERASMUS

230 Joseph Collen tine

programs that explicitly purport to promote learners' proficiency. We need to learn much more about the effects of SA on advanced learners, especially given that we suspect there is a developmental threshold at which it starts to be generally effective.

Researchers also need to consider the effects of particular teaching strategies and syllabus design on learner development. The knowledge base reviewed above indicates that this area of SA research is still in its descriptive phase. There are good models of curricular design whose impact we should start to investigate.

NOTES

1 See Freed (1995a) for an extensive, historical overview of the research on SA up to 1995.

2 www.kois.go.kr/news /News / newsprint. asp ?serial_no=20050725004. 3 See DuFon and Churchill (2006) for an extensive overview of the research to date on

pragmatic research in the SA context.

REFERENCES

Altbach, P. & Bassett, R (2004). The brain trade. Foreign Policy 44, 30-L Barron, A. (2003). Acquisition in interlanguage pragmatics: Learning how to do things with

words in a study abroad context. Amsterdam: John Benjamins. Batstone, R (2002). Contexts of engagement: A discourse perspective on "intake" and

"pushed output." System 30, 1-14. Biber, D. (1988). Variation across speech and writing. Cambridge: Cambridge University Press. Bick, J. (2007). ArOlmd the campfire, global ghost stories. New York Times, Jlme 2L Brecht, R & Davidson, D. (1991). Language acquisition gains in study abroad: Program

assessment and modification. Paper presented at the NFLC Conference on Language Testing, Washington DC.

Brecht, R, Davidson, D., & Ginsberg, R (1995). Predictors of foreign language gain during study abroad. In B. Freed (ed.), Second language acquisition in a study abroad context (pp. 37-66). Philadelphia: John Benjamins.

Carroll, J. (1967). Foreign language proficiency levels attained by language majors near graduation from college. Foreign Language Annals I, 131-5L

Cheng, A. & Mojica-Diaz, C. (2006). The effects of formal instruction and study abroad on improving proficiency: '"!:'he case of the Spanish subjlmctive. Applied Language Learning 16,17-36.

Chieffo, L. & Zipser, R (2001). Integrating study abroad into the foreign language cur­riculmn. ADFL Bulletin 32, 79-85.

Churchill, E. (2006). Variability in the study abroad classroom and learner competence. In M. DuFon & E. Churchill (eds.), Language learners in study abroad contexts (pp. 203-27). Cleve don, UK: Multilingual Matters.

Cohen, A. & Shively, R (2007). Acquisition of requests and apologies in Spanish and French: Impact of study abroad and strategy-building intervention. Modern Language Journal 91, 189-21.

lCY . We need to learn ecially given that We )e generally effective. r teaching strategies base reviewed above Lve phase. There are start to investigate.

research on SA up to

:5004. he research to date on

30-L how to do things with

ctive on "intake" and

ridge University Press. Times, Jlme 2L

:udy abroad: Program ruerence on Language

language gain during a study abroad context

language majors near

l and study abroad on 7lied Language Learning

foreign language cur-

earner competence. In t contexts (pp. 203-27).

ogies in Spanish and tion. NIodern Language

Study Abroad Research 231

Coleman, J. (1997). Residence abroad within language study. Language Teaching 30, 1-20. J. (2004). The effects of learning contexts on morphosyntactic and lexical devel­

opment. Studies in Second Language Acquisition 26, 227-48. Collentine, J. & Freed, B. (2004). Inh'oduction: Learning context and its effects on second

language acquisition. Studies in Second Language Acquisition 26, 153-72. DeKeyser, R. (1990). From learning to acquisition? Monitoring in the classroom and abroad.

Hispania 73, 238-47. DeKeyser, R. (1991). Foreign language development during a semester abroad. In B. Freed

(ed.), Foreign language acquisition: Research and the classroom (pp. 104-19). Lexington, MA: D. C. Heath.

DeKeyser, R. M. (2001). Automaticity and automatization. In P. Robinson (ed.), Cognition and second language instruction (pp. 125-51). Cambridge, MA: Cambridge Universiti Press.

Dings, A. & Jobe, T. (2003). Negotiating obstacles: Extended repair sequences in native I non native interaction. In P. Kempchinsky & c. Pineros (eds.), Theory, practice, and acquisition: Papers from the sixth Hispanic Linguistics Symposium and the fifth Conference on the Acquisition of Spanish and Portuguese (pp. 371-81). Somerville, MA: Cascadilla.

Doughty, C. & Long, M. (2003). Optimal psycholinguistic environments for distance foreign language learning. Language Learning and Technology 7, 50-80.

DuFon, M. & Churchill, E. (2006). Language learners in study abroad contexts. Clevedon, UK: Multilingual Matters.

Dyson, P. (1988). The year abroad: Report for the Central Bureau for Educational Visits and Exchanges. Oxford: Oxford University Language Teaching Centre.

Freed, B. (1990). Language learning in a study abroad context: The effects of interactive and non interactive out-of-class contact on grammatical achievement and oral pro­ficiency. In J. Alatis (ed.), Linguistics, language teaching and language acquisition: The interdependence of theory, practice, and research (pp. 459-77), Georgetown University rolmd table on languages and linguistics. Washington, DC: Georgetown University Press.

Freed, B. (1995a). Introduction. In B. Freed (ed.), Second language acquisition in a study abroad context (pp. 3-34). John Benjamins: Philadelphia.

Freed, B. (1995b). Second language acquisition in a study abroad context. John Benjamins: Philadelphia.

Golonka, E. (2006). Predictors revised: Linguistic knowledge and metalinguistic aware­ness in second language gain in Russian. Modem Language Journal 90, 496-505.

Gore, J. (2005). Dominant beliefs and alternative voices: Discourse, belief, and gender in american study abroad. New York: Routledge.

Guntermann, G. (1995). The peace corps experience: Language learning in training and in the field. In B. Freed (ed.), Second language acquisition in a study abroad context (pp. 149-70). John Benjamins: Philadelphia.

Hoffman-Hicks, S. (1999). The longitudinal development of French foreign language prag­matic competence: Evidence from study abroad participants. Unpublished doctoral dissertation, Indiana University.

Howard, M. (2002). Prototypical and non-prototypical marking in the advanced learner's aspectuo-temporal system. EuroSLA Yearbook 2,87-114.

Howard, M. (2006). Variation in advanced French interlanguage: A comparison of three (socio)linguistic variables. Canadian Modern Language Review 62, 379-400.

Howard, M., Lemee, 1., & Regan, V. (2006). The L2 acquisition of a socio-phonetic variable: The case of III deletion in French. Journal of French Language Studies 16, 1-24.

232 Joseph Collen tine

Isabelli, C. & Nishida, C. (2005). Development of the Spanish subjunctive in a nine-month shJdy-abroad setting. In D. Eddington (ed.), Selected proceedings of the sixth Conference on the Acquisition of Spanish and Portuguese As First and Second Languages (pp. 78-91). Sommerville, MA: Cascadilla Press.

Izumi, S. & Iwasaki, M. (2004). Development of explicit and implicit knowledge of dif­ferent grammatical forms by learners with different ESL learning backgrOl.illds. Sophia Unguistica 52, 1-33.

Kinginger, C. (2007). Research on language education through programmes of study and residence abroad: Recent publications. International Journal of Bilingual Education and Bilingualism 10, 104-12.

Kinginger, C. (in press). Language learning in study abroad: Case shldies of Americans in France. lVlodern Language Journal.

Lafford, B. (1995). Getting into, through, and out of a survival sihlation: A comparison of communicative strategies used by shldents studying Spanish-abroad and Uat home." In B. Freed (ed.), Second language acquisition in a study abroad context (pp. 97-122). John Benjamins: Philadelphia.

Lafford, B. (2004). The effect of context of learning on the use of communication strategies by learners of Spanish as a second language. Studies in Second Language Acquisition 26, 201-26.

Lafford, B. & Collentine, J. (2006). The effects of study abroad and classroom contexts on the acquisition of Spanish as a second language: From research to application. In B. Laf£ord & R. Salaberry (eds.), Spanish second language acquisition: From research findings to teaching applications (pp. 103-26). Washington, DC: Georgetown University Press.

Lazar, N. (2004). A short survey on causal inference, with implications for context of learning studies of second language acquisition. Studies in Second Language Acquisition 26,329-48.

Mackey, A. & Goo, J. (2007). Interaction research in SLA: A meta-analysis and research synthesis. In A. Mackey (ed.), Conversational interaction in second language acquisition (pp. 407-52). Oxford: Oxford University Press.

Magnan, S. & Back, M. (2007). Social interaction and linguistic gain during Shldy abroad. Foreign Language Annals 40, 43-61.

McGeeking, A. (2006). Negotiation in a Japanese Shldy abroad setting. In M. DuFon & E. Churchill (eds.), Language learners in study abroad contexts (pp. 177-202). Clevedon, UK: Multilingual Matters.

Mahle, D. & Raupach, M. (1983). Plan en in der Fremdsprach. Frankfurt: Peter Lang. Mmphy-Lejelme, E. (2002). Student mobility and narrative in Europe: The new strangers.

New York: Routledge. Norris, J. & Ortega, L. (2001). Does type of insh'uction make a difference? Substantive

findings from a meta-analytic review. Language Learning 51, 157-213. O'Brien, L, Segalowitz, N., Collentine, L & Freed, B. (2006). Phonological memory and

lexical, narrative, and grammatical skills in second-language oral production by adult learners. Applied Psycho linguistics 27, 377-402.

O'Brien, L, Segalowitz, N., Freed, B., & Collentine, J. (2007). Phonological memory predicts second language oral fluency gains in adults. Studies in Second Language Acquisition 29, 557-81.

Opper, S., Teichler, u., & Carlson, J. (eds.) (1990). Impact of study abroad programmes on students and graduates. London: Jessica Kingsley.

unctive in. a nine-month !gs of the sixth Conference d Languages (pp. 78-91).

plicit knowledge of dif­ing backgrOl.illds. Sophia

ogrammes of Shldy and F Bilingual Education and

studies of Americans in

. lation: A comparison of -abroad and "at home." mtext (pp. 97-122). John

use of commlmication dies in Second Language

and classroom contexts research to application. 'cquisition: From research Georgetown University

lications for context of md Language Acquisition

a-analysis and research :ond language acquisition

n during shlCly abroad.

.etting. In M. DuFon & JP· 177-202). Clevedon,

furt: Peter Lang. 'ope: The new strangers.

difference? Substantive 57-213. >l1ological memory and ral production by adult

ogical memory predicts Language Acquisition 29,

lj abroad programmes on

Study Abroad Research 233

Pellegrino-Aveni, V. (2005). Study abroad and second language use: Constructing the self. Cam­bridge: Cambridge University Press.

Perdue, C. & Klein, W. (1992). Why does the production of some learners not gram­maticalize? Studies in Second Language Acquisition 14,259-72.

Regan, V. (1995). The acquisition of sociolinguistic native speech norms: Effects of a year abroad on second language learners of French. In B. Freed (ed.), Second language acquisition in a study abroad context (pp. 245-68). John Benjamins: Philadelphia.

Rivers, W. (1998). Is being there enough? The effects of home stay placements on language gain during study abroad. Foreign Language Annals 31, 492-500.

Rodriguez, S. (2001). The perception of requests in Spanish by instructed learners of Span­ish in second- and foreign-language contexts: A longitudinal study of acquisition patterns. Unpublished doctoral dissertation, Indiana University .

Segalowitz, N. (2003). Automaticity and second languages. In C. J. Doughty and M. H. Long (eds.), Handbook of second language acquisition (pp. 382-408). Malden, MA: Blackwell.

Segalowitz, N. & Freed, B. (2004). Context, contact, and cognition in oral Huency acquisi­tion. Studies in Second Language Acqllisition 26, 173-99.

Segalowitz, N., Freed, B., Collentine, J., Lafford, B., Lazar, N., & Dfaz-Campos, M. (2005). A comparison of acquisition of Spanish as a second language in two different contexts of learning: Shldy Abroad and the versus regular academic classroom. Frontiers 10, 1-18.

Shardakova, M. (2005). Interculhual pragmatics in the speech of American L2 learners of Russian: Apologies offered by Americans in Russian. Intercultural Pragmatics 2, 423-51.

Siegal, M. (1995). Individual differences and Shldy abroad: Women learning Japanese in Japan. In B. Freed (ed.), Second language acquisition in a study abroad context (pp. 225-44). John Benjamins: Philadelphia.

Smartt, J. & Scudder, R. (2004). Immersion study abroad in Mexico: Using repair behaviors to assess proficiency change. Foreign Language Annals 37, 592-607.

Spada, N. (1985). Effects of informal contact on classroom learners' proficiency: A review of five studies. TESL Canada Journal 2, 51-62.

Spada, N. (1986). The interaction between types of contact and types of instruction: Some effects on the second language proficiency of adult learners. Studies in Second Lan­guage Acquisition 8, 181-99.

Tanaka, K. & Ellis, R. (2003). Study-abroad, language proficiency, and learner beliefs about language learning. JALT Journal 25, 63-85.

Tiec1uer, U. & Steube, W. (1991). The logics of Shldy abroad programs and the impacts. Higher Education 21, 325-49.

Tokowicz, N., Michael, E., & Kroll, J. (2004). The roles of study-abroad experience and working-memory capacity in the types of errors made during translation. Bilin­gualism: Language and Cognition 7, 255-72.

Wilkinson, S. (1998). On the nature of immersion during study abroad: Some participants' perspectives. Frontiers 4, 121-38.

Wilkinson, S. (2002). The omnipresent classroom durul.g summer study abroad: American students in conversation with their French hosts. Modern Language Journal 86, 157-73.

Willis, F., Doble, G., Sankarayya, u., & Smithers, A. (1977). Residence abroad and the student of modern languages: A preliminary survey. Bradford, UK: Modern Languages Centre, University of Bradford.